This material is the copyright of the original publisher.

Unauthorised copying and distribution is prohibited.

2008, Vol. 1, No. 3 (pp. 211-222)

ISSN: 1178-1653

Original Research Article

Patients Time for Outpatient Visits

Terms and Conditions for Use of PDF

The provision of PDFs for authors' personal use is subject to the following Terms & Conditions:
The PDF provided is protected by copyright. All rights not specifically granted in these Terms & Conditions are expressly
reserved. Printing and storage is for scholarly research and educational and personal use. Any copyright or other notices
or disclaimers must not be removed, obscured or modified. The PDF may not be posted on an open-access website
(including personal and university sites).
The PDF may be used as follows:
to make copies of the article for your own personal use, including for your own classroom teaching use (this includes
posting on a closed website for exclusive use by course students);
to make copies and distribute copies (including through e-mail) of the article to research colleagues, for the personal use
by such colleagues (but not commercially or systematically, e.g. via an e-mail list or list serve);
to present the article at a meeting or conference and to distribute copies of such paper or article to the delegates
attending the meeting;
to include the article in full or in part in a thesis or dissertation (provided that this is not to be published commercially).

ORIGINAL RESEARCH ARTICLE

Patient 2008; 1 (3): 211-222

1178-1653/08/0003-0211/$48.00/0
2008 Adis Data Information BV. All rights reserved.

How Much Time Do Patients

Spend on Outpatient Visits?
The American Time Use Survey
Louise B. Russell,1 Yoko Ibuka1 and Deborah Carr2
1
2

Institute for Health, Health Care Policy, and Aging Research and Department of Economics,
Rutgers University, New Brunswick, New Jersey, USA
Institute for Health, Health Care Policy, and Aging Research and Department of Sociology,
Rutgers University, New Brunswick, New Jersey, USA

Abstract

Background: In Crossing the Quality Chasm, the Institute of Medicine recommended that patient-centered care should not waste patients time and should
recognize the involvement of family and friends. Studies have documented the
time spent by physicians on outpatient visits, but not that spent by patients and
their companions. The patients perspective provides an important yet overlooked
indicator of healthcare effectiveness.
Objective: To document how much time American patients spend on outpatient
visits, for what purposes (travel, waiting, receiving services), and the time
required of family members and friends.
Methods: We used data from the first 4 years (20036) of the American Time Use
Survey (ATUS), conducted by the US Census Bureau for the Bureau of Labor
Statistics, which asks respondents about their activities over a 24-hour period.
ATUS is a nationally representative population-based survey that samples days
continuously throughout the year. In 20036, 60 674 respondents aged 15 years
were randomly selected from households that completed the Current Population
Survey; 1621 reported seeking medical care for themselves on their survey day.
We documented the percentage of the population that reported outpatient visits,
the percentage who were accompanied to those visits and by whom, and the mean
time spent by patients and their companions, by type of activity, and by age and
sex.
Results: After weighting the data to represent the US population, we found that
3.4% of people aged 15 years reported traveling, waiting, or receiving services in
connection with an outpatient visit on their survey day. The mean time for those
who reported the activity was 35 minutes for travel (95% CI 33, 37), 42 minutes of
waiting (95% CI 37, 47), and 74 minutes receiving services (95% CI 70, 79).
Overall, 39.5% were accompanied, usually by family members. Companions
spent a mean of 124 minutes per encounter (95% CI 112, 135). Nearly half of
those aged 65 years were accompanied, almost always by adults only, suggesting
that they may have needed help with transportation, negotiating the healthcare
system, or performing cognitive and emotional tasks involved in receiving care.

Russell et al.

212

Conclusion: Outpatient visits are time intensive for American patients and their
families: the equivalent of 207 million 40-hour work-weeks each year. Patients
and their families spend substantially more time on outpatient visits than the time
with the physician reported by the annual National Ambulatory Medical Care
Survey. Further research is needed on the components of outpatient visits that do
not directly involve physicians. Efforts to improve care should address waiting
times and recognize the involvement of family members. The ATUS could
provide periodic benchmarks of patient time use as a supplement to other
indicators of patient-centered care in the annual National Healthcare Quality
Report.

Background
In its landmark report, Crossing the Quality
Chasm, the Institute of Medicine proposed six aims
for a better health system.[1] Care should be safe,
effective, patient-centered, timely, efficient, and equitable. The report recommended that patient-centered care be available in diverse forms not just
face-to-face visits; that patients receive the information needed to make their own choices; that the
system not waste resources or patient time; and that
it recognize the involvement of family and friends.
To achieve these aims, the US healthcare system
needs a wide range of information about patients
and their families, including their circumstances,
preferences, and experiences in the system.
The American Time Use Survey (ATUS), a new
survey from the US Bureau of Labor Statistics[2] that
documents how people spend their time, offers an
innovative perspective on patients experiences. For
a nationally representative sample of non-institutionalized civilian adults aged 15 years, and using a
24-hour time diary, the gold standard for measuring
time use,[3] the ATUS documents who reported outpatient visits; the amount of time they spent traveling, waiting, and receiving services; and who accompanied them. An analysis of all health-related
activities in the ATUS, based on the first 2 survey
years (20034), showed that 3.4% of the population
received medical care on their survey day and spent
a mean of 123 minutes per person receiving such
care.[4]
In this article, we present data for 20036, the
first 4 survey years. The large sample allows de 2008 Adis Data Information BV. All rights reserved.

tailed examination of time spent traveling, waiting,

and receiving services, and how often family or
friends accompanied the patient. Previous studies
have reported the time spent by physicians on outpatient visits.[5-9] The ATUS, by contrast, provides the
only nationally representative record of patients
perceptions of the time they spend on such visits.
These data shed light on how efficiently the health
system uses patients time and on the involvement
of family members, suggesting a new perspective on
what it means for care to be patient centered.
Methods
The ATUS, conducted by the US Census Bureau
for the Bureau of Labor Statistics, is designed to
produce nationally representative estimates of how
people spend their time.[10,11] Households are selected from those that complete their final interview
for the Current Population Survey, the nations
monthly labor force survey. After the Current Population Surveys oversampling of small states is
corrected, households are stratified by race and Hispanic origin, presence and age of children, and, for
childless households, number of adults, and sampled
at different rates within each stratum. An individual
respondent is randomly selected from people aged
15 years in each household. In 2003, 3375 households were selected each month. In 20046, the
number was reduced to 2194 households per month
for budgetary reasons.
Each months sample is partitioned into four
subgroups; one for each week of the month. Within
each week, 10% of the sample is assigned to each
Patient 2008; 1 (3)

Patients Time for Outpatient Visits

weekday, and 25% to each weekend day. The ATUS

sample weights adjust for this oversampling of
weekend days as well as for differential rates of nonresponse. Response rates declined slightly from
57.8% in 2003 to 55.1% in 2006.[10,12]
The survey is conducted using computer-assisted
telephone interviewing (CATI). Respondents are
randomly assigned a day of the week and phoned the
next day. If interviewers do not reach the respondent, they attempt subsequent contacts on the same
day of the week for up to 8 consecutive weeks. The
5% of households that do not provide telephone
numbers are mailed a request to call the telephone
center for the interview.
During the interview, respondents are asked how
they spent the 24 hours beginning 4:00 am the
previous day (their designated day) and ending
4:00 am the day of the call. Except for common
activities such as eating and sleeping, interviewers
record respondents verbatim descriptions. For activities other than personal care, respondents are
asked where they were and who was in the room or
accompanied them. Responses are coded independently by two interviewers who did not conduct the
interview; coding differences are resolved by
trained adjudicators.[10] Each activity is assigned a
6-digit code; the first two digits indicate one of 17

213

major activity categories, the next four signify an

intermediate category and specific activity.[13,14] The
ATUS data file shows the times that each activity
began and ended.
Table I lists the activities for which we report
data in this paper: activity type, ATUS title and
codes, and examples of activities falling under that
code.[14] These codes represent care that the respondent sought for himself or herself; different codes
are used for time spent accompanying others as they
seek care. As the examples show, all types of outpatient visits are included in the ATUS not just visits
to physicians, but visits to all outpatient healthcare
providers. However, the ATUS does not identify the
specific type of provider seen during the outpatient
visit.
To provide a descriptive account of the type of
care sought, we supplemented our analysis with data
obtained from another survey, the US Medical Expenditure Panel Survey (MEPS),[15] conducted by
the Agency for Healthcare Research and Quality.
Although the two surveys are unrelated, the MEPS
defines the universe of outpatient visits broadly, as
does the ATUS. We extracted data from the 2004
outpatient event files of the MEPS, and, using the
MEPS sample weights, calculated the population

Table I. Outpatient visits: activities and codes in the American Time Use Survey (ATUS), 20036, with examples[14] a
Activity, ATUS title (codes)

Examples

Travel
Travel related to using medical services
2003 and 2004 (170804)

None provided by ATUS

2005 and 2006 (180804)

None provided by ATUS

Waiting
Waiting associated with medical services (080403)

Waiting at the physicians office; waiting for lab test results; waiting
for surgery

Receiving services
Using health and care services outside the home (080401)

Having a doctors appointment, a physical, dental work, eye exam,

inpatient/outpatient treatment, physical therapy, seeing other
healthcare practitioners
Purchasing/paying for healthcare services, including elder care
services
Talking to/with a doctor, nurse, physical therapist, psychologist,
pharmacist, or other healthcare practitioner

Using medical services, not elsewhere classified (080499)

None provided by ATUS
a Using in-home health and care services (080402) was excluded from our analysis.

2008 Adis Data Information BV. All rights reserved.

Patient 2008; 1 (3)

Russell et al.

214

Table II. Number of respondents who reported an outpatient visit,

by age group and sex, 20036 American Time Use Surveya
Age (y)

Men

1544

163

Women
398

Total
561

4564

186

401

587

65

180

293

473

Total
529
1092
1621
a Unweighted. Total respondents, 20036: 60 674.

shares of visits accounted for by major provider

types, which we report in the results.
In the 20036 ATUS, 1621 of 60 674 respondents reported outpatient visits in which they sought
care for themselves (table II). The time pattern of
activities suggested that 139 respondents (8.6%)
might have had two distinct visits on the same day;
the time for both visits was combined in all calculations. Although inpatient stays are included in the
ATUS activity definitions, no respondents reported
times long enough to suggest that they were inpatients on their survey day.
Each person who accompanied the respondent is
assigned an ATUS code specifying his or her relationship to the respondent, and whether she/he was
an adult or a child aged <18 years. In our analysis of
respondents companions, we excluded other nonhousehold adult aged 18. Our preliminary analyses indicated that this category includes the clinician
present at the visit.
In addition to adjusting for the oversampling
of weekend days and for non-response, the ATUS
sample weights (TU06FWGT for 20035,
TUFINLWGT for 2006) allow survey years to be
combined.[11] Confidence intervals were defined using standard errors calculated by the replicate variance method, which accounts for the increase in
variance associated with clustering in the ATUS
relative to the variance expected in a simple random
sample of the same size.[16,17] We used the
DESCRIPT procedure in SAS-callable SUDAAN,
with the replicate weights provided by ATUS, to
compute the standard errors.
The response rate for the ATUS has been just
under 60% instead of the 70% envisioned when the
survey was being developed.[10,11] Three design features contribute to the lower response rate:[18] (i) the
2008 Adis Data Information BV. All rights reserved.

sample is drawn from households that have participated in the Current Population Survey for 8 months
(survey fatigue is the most frequent reason for refusing the ATUS); (ii) only the person selected, not a
proxy, can complete the survey; and (iii) the designated day of the week cannot be changed, although
interviewers can try the same day in subsequent
weeks.
An analysis of the 2004 ATUS[12] found that nonresponse was more often due to inability to reach
respondents, despite valid contact information
(60%), than to refusals (40%). Busy people (proxied
by work hours and children in the household) were
as likely to respond as those less busy, but socially
isolated people (indicated by marital status, schoolage children, homeownership, etc.) had lower response rates. Response rates were lower for men than
women, and for those aged 1545 years than older
people. The authors of the analysis applied three
alternative adjustments for differential response. All
three produced similar estimates of mean times devoted to specific activities, evidence that differential
response rates did not bias the results. While the
analysis did not find significant bias, the authors
could not ascertain whether non-response was related to health, because the ATUS did not ask about
health. It is reassuring that the response rate is
higher for older than younger people, but those of
any age with serious health problems may be less
likely to respond. Thus, the ATUS may be best
suited for describing routine outpatient visits, rather
than more intensive healthcare use. The 2006 survey
included the first health data (self-reported health
status, weight, and height released in June 2008),
which will permit analysis of response rates by
health status.
Data, sample weights, replicate weights, questionnaires, and users guides for each year are available at the US Bureau of Labor Statistics website.[2]
Each years data are available in an activity file,
which records activities in order of occurrence, and
an activity summary file, which reports total time
spent by the respondent on each 6-digit activity.
Respondent and household characteristics are stored
in the ATUS-CPS, respondent, and roster files.
Patient 2008; 1 (3)

Patients Time for Outpatient Visits

215

Our research did not require Institutional Review

Board approval since all the data are publicly available: the ATUS at the Bureau of Labor Statistics
website;[2] and the MEPS at the Agency for Healthcare Research and Qualitys website.[15]
Results
When respondents were weighted to reflect the
non-institutionalized civilian US population, 3.4%
of US adults aged 15 years reported an outpatient
visit on their survey day in the years 20036.
Women aged <65 years were more likely than men
of the same age to report a visit: 3.0% of women
aged 1544 years (95% CI 2.6, 3.4), compared with
1.6% of men (95% CI 1.3, 1.9), and 4.7% of women
aged 4564 years (95% CI 4.1, 5.2) versus 3.0% of
men (95% CI 2.5, 3.5). Rates rose with age to 6.4%
for women aged 65 years (95% CI 5.6, 7.2), and
6.1% for men (95% CI 5.1, 7.1).

Since the ATUS does not identify the type of

provider seen during a visit, we turned to the MEPS
(see the Methods section), which defines outpatient
visits in a similarly broad fashion, for more information. Our calculations based on the 2004 MEPS
show that 47.8% of outpatient visits were to officebased physicians. Visits to hospital outpatient departments accounted for 7.3%, emergency rooms for
2.8%. The largest shares of other office-based
medical providers were dentists (15.4%), chiropractors (5.4%), physical/occupational therapists
(4.2%), and nurses/nurse practitioners (3.8%). All
other providers accounted for 13.3% of the total
visits.
Time Traveling, Waiting, Receiving Services

Nearly all (96%) people with an outpatient visit

reported traveling and/or receiving services as part
of the encounter (table III). One-third (34.2%) also
reported waiting. Mean travel time for those who

Table III. People who reported an outpatient visit:a mean (95% CI) time per person who reported the activity and percentage who reported
the activity, by activity, age, and sex, 20036 American Time Use Surveyb
Activity/Respondent

Men

age (y)

minutes/day

Women
percentage

minutes/day

Total
percentage

minutes/day

percentage

Travel to and from services

1544

32 (27, 37)

94.5

34 (30, 38)

95.3

34 (30, 37)

4564

41 (36, 47)

97.2

35 (31, 40)

96.4

38 (34, 41)

95.0
96.7

65

37 (32, 42)

97.5

33 (29, 37)

96.9

35 (31, 38)

97.1

Total

37 (34, 40)

96.4

34 (32, 37)

96.1

35 (33, 37)

96.2

1544

30 (18, 42)

34.5

44 (31, 57)

32.3

39 (29, 48)

33.1

4564

38 (29, 48)

29.6

43 (34, 51)

33.2

41 (35, 48)

31.9

65

48 (34, 62)

34.5

47 (35, 59)

41.8

47 (39, 56)

38.8

Total

38 (31, 45)

32.7

44 (38, 51)

35.1

42 (37, 47)

34.2

1544

86 (71, 101)

99.2

67 (59, 75)

97.7

73 (66, 81)

98.2

4564

84 (71, 97)

98.8

72 (65, 80)

98.8

77 (70, 84)

98.8

65

86 (69, 103)

96.6

62 (55, 70)

97.7

72 (64, 80)

97.3

Total

85 (76, 94)

98.3

68 (63, 72)

98.1

74 (70, 79)

98.2

Waiting for services

Receiving services

All activities
1544

126 (108, 143)

100

112 (101, 123)

100

117 (107, 126)

100

4564

135 (121, 149)

100

120 (110, 130)

100

125 (117, 134)

100

65

136 (117, 154)

100

112 (102, 122)

100

122 (112, 132)

100

Total
132 (122, 143)
100
115 (108, 121)
100
121 (116, 127)
100
a Percentage of respondents receiving services is <100 as some did not actually receive services, for a variety of reasons.
b

Weighted to reflect each respondents share of the non-institutionalized civilian population aged 15 years.

2008 Adis Data Information BV. All rights reserved.

Patient 2008; 1 (3)

Russell et al.

216

traveled was 35 minutes, mean waiting time for

those who waited was 42 minutes, and mean time
receiving services was 74 minutes. Mean total time
was 121 minutes. Median times were somewhat
lower: 30 minutes traveling, 27 minutes waiting,
60 minutes receiving services, and 100 minutes total
time.
The percentage who reported waiting (34.2%)
seemed low, so we speculated that waiting time may
be under-reported. Waiting may be such a common
component of the healthcare process that patients,
especially those who wait only briefly, take it for
granted and do not always report it as a discrete part
of the visit. To explore this possibility, we compared
total time spent waiting and receiving services for
people who reported waiting and those who did not.
Those who reported waiting averaged 105 minutes
in total: 40 minutes waiting and 65 minutes receiving services. The mean for those who only reported
receiving services was 79 minutes. The 14-minute
difference in time receiving services may be a reasonable estimate of waiting time for these people.
Involvement of Family and Friends

Almost 40% of people with outpatient visits reported that someone accompanied them (table IV;
the 1544 age group is subdivided here since adolescents and young adults may still be accompanied by
parents). Those aged <25 years and those aged
65 years were most likely to be accompanied
(54.6% and 48.5%, respectively). The majority of
companions (85%) were family members.
Most of those accompanied (31.5% of all patients
with outpatient visits) were accompanied only by
other adults; 4.5% were accompanied only by children, and 3.4% were accompanied by both children

and adults (table V). Women were more likely than

men to be accompanied by children only (data not
shown).
Companions spent a substantial amount of time
in accompanying respondents to outpatient visits.
Table V shows companions time per person with an
outpatient visit. The total amount of companions
time can be longer than patients time because onefifth of patients who were accompanied were accompanied by two or more people. In total, per
person seeking care, companions devoted 123 minutes (children only), 235 minutes (children and
adults), and 112 minutes (adults only).
People who spent 90 minutes receiving services, who may have been undergoing extensive
testing or outpatient procedures, were more likely to
be accompanied (46.7% vs 36.7%; p < 0.001) and
more likely to be accompanied only by other adults
(39.4% vs 28.5%; p < 0.001).
Discussion
Patients and their families spend a substantial
amount of time seeking medical care. Data for
20036 from the nationally representative ATUS
show that, on average, 3.4% of US adults aged
15 years had an outpatient visit each day, which
required a mean of 2 hours traveling, waiting, and
receiving services. Almost 40% of patients were
accompanied, usually by family members, who
spent another 2 hours.
According to Consumer Reports, patients most
common complaint about doctors is that they were
kept waiting: in that 2006 survey, 24% reported that
they waited 30 minutes.[19] Of ATUS respondents
who reported outpatient clinical encounters (a
broader category than people visiting the doctor),

Table IV. Percentage (95% CI) of people with an outpatient visit who were accompanied, by age and sex, 20036 American Time Use
Surveya
Age (y)

Men

Women

Total

1524

45.4 (24.3, 66.6)

61.1 (46.1, 76.0)

54.6 (40.8, 68.4)

2544

32.7 (21.9, 43.5)

39.0 (32.3, 45.7)

37.0 (31.8, 42.2)

4564

27.0 (19.8, 34.3)

31.8 (26.2, 37.4)

30.0 (25.5, 34.5)

65

48.3 (39.7, 56.9)

48.6 (41.3, 56.0)

48.5 (42.8, 54.2)

Total
36.9 (31.8, 42.0)
41.0 (37.2, 44.8)
39.5 (36.4, 42.6)
a Weighted to reflect each respondents share of the non-institutionalized civilian population aged 15 years.

2008 Adis Data Information BV. All rights reserved.

Patient 2008; 1 (3)

Patients Time for Outpatient Visits

217

Table V. Percentage of people accompanied, time spent, and percentage who waited, by type of companion and age of respondent,
20036 American Time Use Surveya
Respondents age
(y)

Percentage of
age group

Respondents total
mean time (95% CI)

Companions total
mean timeb (95% CI)

Percentage who reported

waiting (95% CI)

1524

45.4

106 (77, 134)

37.2 (15.5, 58.9)

2544

63.0

110 (99, 120)

27.8 (22.2, 33.4)

4564

70.0

111 (103, 119)

28.5 (22.5, 34.6)

65

51.5

108 (97, 119)

32.5 (25.7, 39.2)

Total

60.5

109 (103, 115)

29.9 (26.1, 33.7)

Not accompanied

Accompanied by children only (<18 y)

1524c
2544

11.0

104 (79, 129)

122 (82, 162)

39.9 (25.8, 54.1)

4564

2.3

112 (69, 155)

119 (80, 157)

30.6 (1.2, 60.0)

4.5

107 (87, 127)

123 (89, 157)

36.5 (25.2, 47.7)

65c
Total

Accompanied by children (<18 y) and adults

1524c
2544

8.4

136 (96, 176)

261 (184, 338)

30.7 (15.1, 46.4)

4564

1.6

195 (107, 283)

321 (141, 501)

36.0 (5.1, 66.9)

3.4

145 (113, 177)

235 (171, 299)

32.5 (19.7, 45.3)

65c
Total

Accompanied by adults only

1524

45.5

142 (107, 177)

121 (90, 152)

43.0 (24.8, 61.2)

2544

17.6

132 (108, 156)

106 (82, 130)

35.6 (23.7, 47.5)

4564

26.1

162 (141, 183)

126 (102, 149)

40.6 (30.5, 50.7)

65

46.6

135 (119, 152)

101 (82, 120)

46.0 (38.2, 53.8)

Total

31.5

143 (132, 154)

112 (100, 124)

42.4 (37.8, 47.1)

Overall total
39.5
121 (116, 127)
124 (112, 135)
34.2 (31.3, 37.2)
a Weighted to reflect each respondents share of the non-institutionalized civilian population aged 15 years.
b

Mean per respondent, not per companion.

Numbers based on fewer than ten respondents have been omitted.

34% said they waited. About half of those (17.1% of

all those with visits) waited 30 minutes. Mean
waiting time for those who reported waiting was
42 minutes. Waiting may be under-reported: we
estimated that respondents who did not report waiting separately may have averaged 14 minutes of
waiting.
National Level Estimates

Counting patients and companions time, the

ATUS data show that 35 hours almost a full
workweek was spent annually on outpatient visits
for each person in the population aged 15 years.
Annual time for patients was derived by multiplying
2008 Adis Data Information BV. All rights reserved.

the proportion of people who report visits (0.034) by

the mean time spent (121 minutes, table III), yielding 4.1 minutes per person per day, or 25 hours per
year. Annual time for companions was derived by
multiplying the proportion who reported visits
(0.034) by the proportion who were accompanied
(0.395, table IV) and the mean time spent by companions (124 minutes, table V), adding another
1.7 minutes per person per day, or 10 hours per year.
Not everyone spent this much time; some spent less,
others especially elderly people spent more.
Multiplying 35 hours per person by the 2005
population aged 15 years, patients and their companions spent a total of 207 million 40-hour workweeks on outpatient visits each year. This number
Patient 2008; 1 (3)

Russell et al.

218

does not include time spent by children aged

<15 years, who are not included in the ATUS, and
their companions. The total makes it clear that, at
both the individual and national levels, outpatient
visits represent a substantial commitment of time on
the part of patients and their families.
Comparison with Time Spent by Physicians

The National Ambulatory Medical Care Survey,

and special surveys, show that on average physicians spend <20 minutes face to face with patients.[5-9] The time that patients reported for receiving services was considerably longer a mean of
74 minutes and a median of 60 minutes. Mean time
is increased by about 5 minutes because we summed
times for the 8.6% of the sample whose activities
suggested they might have had two separate visits
on their survey day; the median is unaffected. The
ATUS uses 24-hour time diaries, which is the gold
standard for collecting time-use data because it constrains respondents to fit the times for individual
activities into the 24-hour total;[3] thus the times are
not likely to be overestimated. Rather, they are very
similar to mean times reported in a study of a
hospital-based outpatient teaching clinic, where patients who did not use an interpreter spent 82 minutes in the clinic, 28 of them with the provider,
compared with 100 and 36 minutes for those who
used an interpreter.[20] As in the National Ambulatory Medical Care Survey, time receiving services
differed little by age.[6]
The main reason for the difference between reports of time use by physicians and patients is that
an outpatient visit includes many components that
do not directly involve the physician: check-in,
which can require completing short forms for returning patients and longer forms for new patients;
insurance verification; the trip to the examination
room; time to undress if needed (and dress again
afterward); tests and measures done by staff, such as
height and weight, blood pressure, recording current
symptoms, vision and hearing checks; preparation
for exams such as the Pap smear; having blood
drawn; giving a urine sample; receiving a shot; and
the delays between these tasks. Researchers have
2008 Adis Data Information BV. All rights reserved.

studied the content of physician-patient interactions[5-9,21] and of physicians work outside the
examination room,[7] but have not explored activities that take place when physicians are not present.
Further attention to such activities would shed light
on the different perspectives on outpatient visits.
Our findings underscore that a healthcare encounter
is experienced very differently by physicians and
patients (and their caregivers). Evaluating patients
perceptions and experiences is critical for developing a more thorough understanding of healthcare
delivery, and a more patient-friendly approach to
evidence-based medicine.[22]
Companions and Their Role

The ATUS also reveals an important component

of informal caregiving that has not been noted in the
caregiving literature: accompanying patients to outpatient visits. Traditional studies of caregiving focus
on direct care provided in the home, such as assistance with self-care or bill paying. Our findings
reveal that caregiving extends beyond the boundaries of the patients home. For example, we found
that the mean time devoted by companions was
equal to the time spent by patients themselves. Their
presence and their time suggest that the concept of
patient-centered care may need to be expanded beyond the patient to include companions. They too
need to have their time treated with respect and may
need to be involved in the discussions, exchanges of
information, and decisions that occur in a clinical
encounter.
Approximately half of men and women aged
65 years were accompanied, almost always by
adults only, suggesting that they may have needed
help with transportation, negotiating the healthcare
system, or performing cognitive and emotional tasks
involved in receiving care. People who are deaf or
hard of hearing, for example, bring family members
to help with communication.[23] Older adults with
limited vision, or who are experiencing early symptoms of cognitive decline, may also be particularly
dependent on family members. Crossing the Quality
Chasm noted that the family and friends on whom
patients rely need to feel welcome and comfortable,
Patient 2008; 1 (3)

Patients Time for Outpatient Visits

to be involved in decisions as appropriate, and to

have their needs and contributions recognized.[1]
Bergeson and Dean[24] proposed criteria for a
systems approach to patient-centered care: for example, patients should have access to an appointment when they want or need it and with the clinician they choose. The ATUS shows that for many
patients, appointment schedules must be coordinated with the people who will accompany them. To
the extent that the companions also require medical
attention, it may be helpful to coordinate visits;
older spouses, for example, may benefit from having
visits scheduled at similar times. The multiple
routes of practice access suggested by Bergeson
and Dean[24] (telephone, email, drop-in visits) may
be as important for companions as for patients themselves. Increasing patient participation, coordinating
care across different locations, ensuring that information and support are available and that care reflects the patients needs and interests are other
laudable goals that may need to be expanded to
recognize the involvement of family members, not
just in home care but during clinical encounters. Our
findings underscore the importance of taking a
broad view of Bergeson and Deans recommendation to improve opportunities for patients and families to participate in the care process.
Durso[25] and Peterson[26] focused particularly on
the needs of older adults in a system that strives to
be patient centered. Peterson[26] noted that the complicated tradeoffs involved in treatment decisions
for older patients must be individualized to reflect
their preferences. Writing about diabetes mellitus,
Durso[25] stated that the patients general health
status and goals of care provide important context
when prioritizing and balancing clinical recommendations for older adults. Again, the ATUS data
suggest that these goals need to be expanded to
recognize family members who accompany the elderly patient. Often, that person may be present to
help with the difficult tasks of gathering information, choosing next steps in treatment, and learning
about home-based healthcare regimens such as administering medications or adhering to dietary restrictions.
2008 Adis Data Information BV. All rights reserved.

219

Patients Time as a Starting Point for

Quality-of-Care Measures

Since 2003, the annual National Healthcare

Quality Report, developed and published by the
Agency for Healthcare Quality and Research, has
monitored the timeliness and patient centeredness of
care, among other quality goals.[27] Timeliness is
measured as the percentage of patients who report
that they did not get care for an illness or injury as
soon as they wanted, and the percentage who left an
emergency room without being seen; these data are
supplied by the MEPS and the National Hospital
Ambulatory Medical Care Survey, respectively. Patient centeredness is measured by an index, also
constructed from the MEPS, which shows the percentage of patients who feel that their doctor listens
to them and respects their concerns. As new measures are developed, they are added to the report.
The ATUS could provide new measures to
benchmark timeliness and patient centeredness of
care in the National Healthcare Quality Report.
Two possibilities are the proportion of people who
report waiting more than some length of time, such
as 15 or 30 minutes; and the proportion who report
that someone accompanied them. Both would serve
the agencys goal of informing a partnership
among practitioners, patients, and their families
(when appropriate) to ensure that decisions respect
patients wants, needs, and preferences.
A strength of these two measures is that they
provide a more accurate assessment of the time
spent and with whom, because the ATUS obtains
data within 24 hours of the time the visit occurred.
By contrast, standard measures of satisfaction with
care, often collected long after the encounter, may
be subject to retrospective recall bias.[28] Such measures of satisfaction also may be biased by patients
fondness for their healthcare provider. However,
like many quality (and other) indicators, neither of
the measures we suggest would be a straightforward
indicator of quality. Rather they would serve as
starting points for further investigation on which to
base policy changes.
For waiting time, the ATUS data provide a starting point for exploring how accurately and comPatient 2008; 1 (3)

Russell et al.

220

pletely patients report waiting time and where that

waiting takes place. As suggested earlier, some patients may not report waiting time when it is short.
Furthermore, it is not possible to tell from the ATUS
data whether patients reported only time waiting in
the waiting room, or time waiting in the examination
room as well, and whether they included time waited
because they arrived early for the appointment. Although the ATUS has a procedure for including
supplements to the core survey, it might work better
to explore these issues through supplements to existing health surveys, such as the National Health
Interview Survey (NHIS), the nationally representative health survey that has been conducted by the US
National Center for Health Statistics since 1957.[29]
The supplemental questions could be based on formats often used in the NHIS, such as asking respondents to provide information about outpatient visits
made in the week or 2 weeks preceding the survey. It
would also be important to identify the type of
provider seen, which the ATUS does not do, in order
to focus quality improvement efforts.

it as a measure of opportunity. Wolff and Roter[30]

make the point well in their article, Hidden in Plain
Sight: Medical Visit Companions as a Resource for
Vulnerable Older Adults. Based on a survey of a
representative sample of >12 000 Medicare beneficiaries, they found that the majority of companions
played an active role in the visit asking questions
for the patient, providing information to the physician about the patients condition, and recording and
explaining the physicians instructions for the patient. Thus, they served as a crucial link between
physician and patient. This was particularly the case
for older and more vulnerable patients. Their analysis showed that when companions served in this
role, patients were substantially more satisfied with
the visit. The ATUS data point to this opportunity
for patients of all ages: simply knowing that so many
people are accompanied, and that companions can
serve such an important role, should encourage providers and policy makers to make better use of them.

For policy purposes, the interpretation of information about waiting would need to take into account the larger context of the costs of outpatient
care to patients. Waiting time is only one of the costs
of a visit. Other costs include the time spent traveling and receiving services; the out-of-pocket monetary costs associated with the visit; comfort and
convenience costs such as the ease or difficulty of
dealing with office staff at the visit; childcare or care
for other adults necessary to make the visit possible;
lost income for hourly workers; and so on. In circumstances where out-of-pocket costs for the visit
itself are low or zero, waiting times are one way of
allocating scarce services, so there are trade-offs
between using waiting and monetary costs to limit
the demands on those services. At the same time,
policy makers need to recognize that time costs,
particularly waiting, may discourage patients from
following the ever-increasing number of recommendations for screening and other services.

The time of patients and their families represents

a substantial and valuable resource to the healthcare
system. Researchers have only recently begun to
study this time.[4,31-33] The National Research Councils report, Beyond the Market,[34] recommended
the creation of national accounts to document productive non-market activity; the ATUS would provide data on the unpaid time of patients and caregivers for the proposed National Health Account. The
Public Health Services Panel on Cost-Effectiveness
in Health and Medicine recommended that costeffectiveness analyses include time of patients and
unpaid caregivers in the costs of a health intervention;[35] including that time can change cost-effectiveness ratios substantially.[36,37]

The proportion of patients who are accompanied

is also not a simple quality indicator for which more,
or less time, is unequivocally better. Rather, we see
2008 Adis Data Information BV. All rights reserved.

Patients Time in Economic Analyses

Conclusions
The ATUS documents, for a nationally representative sample of adults, that traveling to outpatient
visits, and waiting for and receiving services, is time
intensive, averaging 2 hours for patients, and, for the
40% who were accompanied, another 2 hours for
their companions. Over a year, for the US populaPatient 2008; 1 (3)

Patients Time for Outpatient Visits

tion aged 15 years, this amounts to 207 million

40-hour work-weeks just for outpatient visits. Inpatient care, personal self-care, including the selfmanagement crucial for chronic diseases, and care
of others requires more time. Although a vast body
of research has documented the financial costs of
care, few studies have documented the time invested
by patients. That time may be taken from paid
employment, or from household tasks, childcare, or
other productive, but unpaid, activities. The time
costs of care, in addition to the financial costs, may
be onerous to many. Our analysis is a first step
toward documenting how much time is invested in
maintaining and caring for ones health and toward
recognizing that cost in a patient-centered system.
Acknowledgments
An abstract based on this paper has been accepted for oral
presentation at the Society for Medical Decision Making
meetings; 2008 Oct 21; Philadelphia (PA), USA.
No sources of funding were used to assist in the preparation of this study. The authors have no conflicts of interest
that are directly relevant to the content of this study.
No one other than the authors made substantial contributions to the work.
Dr Ibuka took up a post-doctoral position at the Yale
School of Public Health in July.

References
1. Committee on Quality of Health Care in America, Institute of
Medicine. Crossing the quality chasm. Washington, DC: National Academy of Sciences Press, 2001
2. US Department of Labor, Bureau of Labor Statistics. American
Time Use Survey [online]. Available from URL: http://www.
bls.gov/tus/home.htm [Accessed 2008 Sep 9]
3. Juster FT, Stafford FP. The allocation of time: empirical findings, behavioral models, and problems of measurement. J Econ
Lit 1991; 29 (2): 471-522
4. Russell LB, Ibuka Y, Abraham KG. Health-related activities in
the American Time Use Survey. Med Care 2007; 45: 680-5
5. Gilchrist VJ, Stange KC, Flocke SA, et al. A comparison of the
National Ambulatory Medical Care Survey (NAMCS) measurement approach with direct observation of outpatient visits.
Med Care 2004; 42: 276-80
6. Lo A, Ryder K, Shorr RI. Relationship between patient age and
duration of physician visit in ambulatory setting: does one size
fit all? J Am Geriatr Soc 2005; 53: 1162-7
7. Gottschalk A, Flocke SA. Time spent in face-to-face patient
care and work outside the examination room. Ann Fam Med
2005; 3: 488-93
8. Cherry DK, Woodwell DA, Rechsteiner EA. National Ambulatory Medical Care Survey: 2005 summary [Advance Data from
Vital and Health Statistics; report no. 387]. Rockville (MD):
National Centre for Health Statistics, 2007 Jun 29

2008 Adis Data Information BV. All rights reserved.

221

9. Tai-Seale M, McGuire TG, Zhang W. Time allocation in primary care office visits. Health Services Res 2007; 42: 1871-94
10. American Time Use Survey Users Guide: Understanding
ATUS 2003 to 2006. Washington, DC: Bureau of Labor Statistics and US Census Bureau, 2007 June [online]. Available
from URL: http://www.bls.gov/tus/atususersguide.pdf [Accessed 2008 Apr 15]
11. Horrigan M, Herz D. Planning, designing, and executing the
BLS American Time-Use Survey. Mon Labor Rev 2004; 127
(10): 3-19
12. Abraham K, Maitland A, Bianchi S. Nonresponse in the American Time Use Survey: who is missing from the data and how
much does it matter? Public Opin Q 2006; 70 (5 Special Issue):
676-703
13. Shelley KJ. Developing the American Time Use Survey activity
classification system. Mon Labor Rev 2005; 128: 3-15
14. US Department of Labor; Bureau of Labor Statistics. American
Time Use Survey: activity coding lexicons. Washington, DC:
Bureau of Labor Statistics [online]. Available from URL:
http://www.bls.gov/tus/lexicons.htm [Accessed 2008 Mar 26]
15. US Department of Health & Human Services, Agency for
Healthcare Research and Quality. Medical Expenditure Panel
Survey (MEPS) [online]. Available from URL: http://www.
ahrq.gov/data/mep six.htm [Accessed 2008 Sep 9]
16. Chapter 14: estimation of variance. In: Design and methodology: current population survey [technical paper 66]. Washington, DC: US Bureau of Labor Statistics; US Census Bureau,
2006 Oct [online]. Available from URL: http://www.cen
sus.gov/prod/2006pubs/tp-66.pdf [Accessed 2007 Oct 5]
17. Fay RE. Theory and application of replicate weighting for
variance calculations. Proceedings of the Survey Research
Methods Section, American Statistical Association, 1989:
212-17 [online]. Available from URL: http://www.amstat.org/
sections/srms/Proceedings/ [Accessed 2008 Sep 9]
18. ONeill GE, Sincavage JR. Response analysis survey: a qualitative look at response and nonresponse in the American Time
Use Survey. Washington, DC: Bureau of Labor Statistics;
2004 [online]. Available from URL: http://www.bls.gov/ore/
pdf/st040140.pdf [Accessed 2007 Oct 5]
19. Get better care from your doctor: what 39 090 patients and 335
doctors have to say about how to make the most of your next
appointment. Consum Rep 2007 Feb; 72 (2): 32-36
20. Fagan MJ, Diaz JA, Reinert SE, et al. Impact of interpretation
method on clinic visit length. J Gen Intern Med 2003; 18:
634-8
21. Stange KC, Zyzanski SJ, Jaen CR, et al. Illuminating the black
box: a description of 4454 patient visits to 138 family physicians. J Fam Practice 1998; 46: 377-89
22. Klag MJ, MacKenzie EJ, Carswell CI, et al. The role of The
Patient in promoting patient-centered outcomes research. Patient 2008; 1 (1): 1-3
23. Iezzoni LI, ODay BL, Killeen M, et al. Communicating about
health care: observations from persons who are deaf or hard of
hearing. Ann Intern Med 2004; 140: 356-62
24. Bergeson SC, Dean JD. A systems approach to patient-centered
care. JAMA 2006; 296: 2848-51
25. Durso SC. Using clinical guidelines designed for older adults
with diabetes mellitus and complex health status. JAMA 2006;
295: 1935-40
26. Peterson ED. Patient-centered cardiac care for the elderly:
TIME for reflection. JAMA 2003; 289: 1157-8

Patient 2008; 1 (3)

222

27. Agency for Healthcare Quality and Research. 2006 National

Healthcare Quality Report [publication 07-0013]. Rockville
(MD): AHRQ, 2006 Dec
28. Grimes D, Schulz K. Bias and causal association in observational research. Lancet 2002; 359: 248-52
29. US National Center for Health Statistics. National Health Interview Survey (NHIS) [online]. Available from URL: http://
www.cdc.gov/nchs/nhis.htm [Accessed 2008 Sep 27]
30. Wolff JL, Roter DL. Hidden in plain sight: medical visit companions as a resource for vulnerable older adults. Arch Int Med
2008; 168 (13): 1409-15
31. Yabroff KR, Warren JL, Knopf K, et al. Estimating patient time
costs associated with colorectal cancer care. Med Care 2005;
43: 640-8
32. Yabroff KR, Davis WW, Lamont EB, et al. Patient time costs
associated with cancer care. J Natl Cancer Inst 2007; 99: 14-23
33. Cantor SB, Levy LB, Cardenas-Turanzas M, et al. Collecting
direct non-health care and time cost data: application to screening and diagnosis of cervical cancer. Med Decis Making 2006;
26: 265-72

2008 Adis Data Information BV. All rights reserved.

Russell et al.

34. Abraham K, Mackie C, editors. Beyond the market. Washington, DC: National Academy of Sciences, 2005
35. Gold MR, Siegel JE, Russell LB, et al., editors. Cost-effectiveness in health and medicine. New York: Oxford University
Press, 1996
36. Jonas DE, Russell LB, Sandler RS, et al. Value of patient time
invested in the colonoscopy screening process: time requirements for colonoscopy study. Med Decis Making 2008; 28:
56-65
37. Lafata JE, Martin SA, Kaatz S, et al. The cost-effectiveness of
different management strategies for patients on chronic warfarin therapy. J Gen Intern Med 2000; 15: 31-7

Correspondence: Professor Louise B. Russell, Institute for

Health, Rutgers University, 30 College Avenue, New
Brunswick, NJ 08901, USA.
E-mail: lrussell@ifh.rutgers.edu

Patient 2008; 1 (3)